Deployment of Software Defined Radio in Space
ITP Capstone Project Research Paper
April 28, 2017 Anisha Teckchandani
Priyanka Balraj
Shivraj Barik
Vidhi Raval
Interdisciplinary Telecom Program
University of Colorado Boulder
Dr. Kevin Gifford
Interdisciplinary Telecom Program
University of Colorado Boulder
Adam M. Schlesinger
Industry Advisor
Abstract- Rapid advancements in terrestrial wireless communications have inspired an engineering assessment of external wireless communication services for the International Space Station (ISS). Our research analyzes several options (802.11n, 802.11ac, LTE, MIMO), and provides recommendations for improving and extending wireless communications in support of increased payload science return and crew operations. Radio frequency propagation modeling of LTE and 802.11n wireless protocols is performed and the results are analyzed. Based on the analysis, a viable protocol is recommended for providing wireless communication in space. The 802.11n and LTE wireless protocols are implemented on a Software Defined Radio (SDR) to establish the feasibility of cognitive radio in space for wireless communication. MATLAB was selected to be the software platform for incorporating the signal propagation between the transmitter and receiver station. Feasibility of Wi-Fi and LTE has been verified with the help of hardware that was selected on the basis of an extensive market research. Based on software simulation using Wireless Insite, LTE has better coverage and capacity in comparison to Wi-Fi on the ISS.
Index Terms—International Space Station (ISS), Long Term Evolution (LTE), IEEE 802.11, Software Defined Radio (SDR), Propagation Modeling, External Wireless Communication (EWC), Wireless Access Points (WAP), Multiple input-multiple-output (MIMO), Wireless Local Area Network (WLAN), Medium-Access-Control (MAC), Physical-Layer (PHY).
I. Research Question and Problem Setting
The International Space Station (ISS) along with six crew members on board is built to support advanced scientific activities including technology research and development in support of National Aeronautics and Space Administration (NASA) exploration activities. There is an inevitable need for
Fig.1. – External surface of ISS
improving wireless communication to support research payload activities and also an increase in scientific data collection and analysis.
Figure 1 illustrates the external surface of the International Space Station. The areas marked with red circles denote the location of External High Definition Camera Assembly (EHDCA) and wireless 802.11n access points mounted on the space station. These cameras have high bandwidth requirements and need a reliable wireless protocol to transmit images/ live feed to a receiver located on the body of the ISS. The current deployment of EWC WAP neither satisfies the capacity nor the coverage requirements of the instruments on the surface of the ISS.
The increase in demand for wireless communication within the internal and external environment of the ISS arises since it serves as a research laboratory and operational work station for crew members. The resulting increase in demand of wireless network communications is similar to terrestrial demand in such spaces. Moreover, the recognized efficiency and automation provided by enhanced wireless technologies can free up the crew from mundane tasks and help facilitate enhanced situational awareness and safety. Research projects and payloads will also benefit directly from the ISS external wireless infrastructure.
In our analysis, radio frequency propagation modeling of LTE and 802.11n wireless protocols is performed to characterize the external RF environment of the ISS. The software platform for designing the protocol architecture on the radio is selected after evaluating the performance of available tools such as GNU Radio, MATLAB and LabVIEW. The current 802.11n network infrastructure on the ISS is compared against optimized Wi-Fi and LTE network designs which includes the use of MIMO and improved node placements. Wireless transmission characteristics evaluated include data rate, mobility (range) and spectral efficiency.
In place of physical access points and eNodeBs an alternative would be to install SDRs. This would enable programmability for the new standard which could be uploaded to the SDR firmware from Earth. To future proof the communications systems for crew members and payloads, this research will propose usage of SDRs that would be adjusted according to the signal path losses in space. The current model of space communication does not incorporate modules which are reconfigurable for future use.
This paper evaluates the existing wireless communications infrastructure external to ISS that is available to research scientists: the current External Wireless Communications (EWC) system is baselined and evaluated with specific recommendations to improve the existing 802.11n based infrastructure. LTE is the suggested improvement to the current 802.11n protocol that is deployed external to the ISS. An LTE enabled wireless communication network, while not currently planned, could be advantageous or even enable certain experiments in addition to the possibility of improved operational benefits.
1. Research Problem:
The project investigates the best way to achieve synchronous wireless communication by implementing 802.11n and LTE protocols for providing reliable transmission of data in space using SDRs.
2. Sub-problems:
A. Selection of RF Modeling Tools
Currently there are various SDRs by different vendors available in the market. Each of these SDRs have defined capabilities and functionalities. For this project, 802.11n and LTE protocols are evaluated. The ideal SDR should be capable of providing dynamic frequency selection for the 802.11n protocol. It should also have the ability to work in the L, Ka, Ku and X band for LTE communication. Various hardware and software platforms were compared to establish a base for the feasibility of implementing Wi-Fi and LTE on SDR.
For RF Propagation Modeling a software tool with capability to enable placement of transmitters and receivers is required. MATLAB, EDX Signal Pro and Ekahau were evaluated and failed to serve this purpose. Wireless InSite was used to generate heat maps and even obtain several other performance parameters.
B. Comparison of Wireless Protocols
The current 802.11n (without MIMO) wireless standard deployed on access points in space fails to provide reliable communication for the payloads and astronauts working on the ISS. It is necessary to implement the most reliable standard that could provide synchronized communication, interference mitigation capabilities and meet the data intensive requirements of the instruments. Wi-Fi and LTE have become the most common wireless technologies that are being used on SDR platforms in the terrestrial environment to enable wireless communication. This research compares these two theoretically as well as with the help of the RF Modeling tool – Wireless InSite, and based on a variety of parameters, recommends the use of LTE in space.
C. RF Propagation Modeling
The RF propagation model for space will differ from the models used on Earth due to factors unique to the space environment, such as scintillation, diffractions and scattering. The space station module can be regarded as an imperfect waveguide. A model is developed using Wireless InSite that is utilized for RF coverage and link performance analysis. Findings from this analysis can be used to determine the appropriate SDR positioning for optimal RF coverage and ultimately to demonstrate the better-performing wireless protocol for the space environment when comparing Wi-Fi and LTE.
II. Literature Review
The current requirements of an SDR include cost reduction, ASIC digital resources, reprogrammable digital resources, advanced digital/ analog converters and API’s [1]. With the right amount of flexibility and programmability, SDRs make a great platform for LTE and Wi-Fi base stations [2]. The Wireless Open-Access Research Platform (WARP), Universal Software Radio Peripheral (USRP) and GNU Radio [3] platforms are potential candidates which will be compared in terms of functionality and efficiency.
[4] states that SDRs are the next generation technology for radio communication. This is primarily because of the interoperability and flexibility that is provided by a single SDR. One SDR can have the capability of supporting features of many different hardware radios. However, this still requires efforts to implement these capabilities using source codes on an SDR.
There has been extensive research in implementing 802.11 on commercial SDRs and on SIMD instructions in a CPU [5] [6]. The MAC proposed in [6] is revolutionary as it entirely implements the physical and virtual carrier sense functions by using RTS/CTS along with the back off process. The SDR developed by Bloessl [6] uses GNU radio features such as stream tagging and message passing using the vectorized library which is based on a frame detection algorithm that auto correlates short training sequences of OFDM frames. They conducted extensive interoperability tests with consumer grade Wi-Fi cards and presented packet delivery ratio measurements to prove that their receivers could decode 20MHz OFDM signals to achieve a reasonable receiver performance [6].
SDRs can be programmed to utilize IEEE 802.11 MAC for developing wireless ad-hoc networking [7]. It emphasizes the usage of SDRs for addressing the challenges of reliable communication and re-configurability of cross-layer communication on a hardware platform [6].
802.11n was deployed using TD-LTE (Time Division Long Term Evolution) on an RTL-SDR using a USRP-2 Board in [8]. The 802.11n protocol, which runs on dual frequency 2.4 GHz and 5GHz was the last wireless protocol to have been implemented and deployed in space on the external walls of the International Space Station. Implementation of LTE on SDRs is still in its nascent stage and mostly focuses on enhancing the processing power of the SDR. In [9], a real-time implementation of LTE with 4X4 MIMO using SDR has been established. The research states that using MIMO and sensing capability, a portable SDR can be developed that would work on speeds that vary from 2 Mbps to 150 Mbps. It also deals with the PHY parameters of the higher layers to conduct an extended experiment on cognitive radio and wireless networking. The major difficulty in mounting LTE on an SDR is due to the LTE base station cell architecture. An LTE transceiver SDR has been developed on a GPU by the Ettus Research Company [10]. Pico LTE is developing a single compact portable box that could provide parallel programming computation between the LTE received signal strength and GPU’s data architecture [10] [11]. The eNodeB implementation along with the downlink/uplink has been successfully implemented on the Amarisoft which uses Matlab and OpenLTE support [12]. But the existence of an OpenBTS is under development, which if developed could provide the base station as well as the user equipment implementation of the protocol [12]. SDRs have been tried and tested on terrestrial and space communication networks.
A major challenge in deploying these radios for enhanced communications is to achieve synchronized transmission and reception of signals. Part of the solution lies in implementing a high efficiency module which can act as a combined processor for LTE and Wi-Fi signals. An additional objective is to facilitate re-configurability by pushing new images of software versions from a terrestrial station with every new release of communication protocols. The aim is to achieve this by building wireless communication (802.11 and LTE) base stations from scratch instead of a plug and play scenario. This brings into picture the need for modification of PHY layers of protocols, more commonly known as Network Coding.
Network coding was initially proposed for wired networks and interoperability in communications [1]. Network Coding of the PHY and MAC layers of wireless networks can fulfil the requirements of schedulers in the MAC layer of the protocol in use [1]. This could be an excellent way to resolve challenges which arise while trying to achieve synchronized wireless communication.
A receiver for communication with satellites was developed on an SDR in the lowest Earth orbit using adaptive OFDM coding [13]. This was achieved using Band Resource Allocation along with 16 QAM Modulation [14]. A clear set distinction is not possible between these platforms due to the various duty cycles used for processing, power/area used for simulation and even the set port accelerators [14] [15]. Better synchronization mechanisms were developed to manage flexible communications between the 802.11n interfaces by using a better carrier tracking algorithm and simpler modulation by applying shorter frame lengths [5]. This could help in setting up synchronization mechanisms for the 802.11n system.
Although a significant amount of work has been previously investigated to design and build transceivers for 802.11n and LTE, none of this was specifically designed for operation in space. Our research aims to simulate the performance of LTE and Wi-Fi communication systems while using RF propagation models of the ionosphere. Furthermore, until now LTE and 802.11 protocols have not been tested against each other. It would be challenging to build on this existing research by implementing both LTE and 802.11n on an SDR and testing them for data rate, throughput and multipath fading around the ISS.
The fundamentals of radio wave propagation in space and experimental views have been reviewed and several prediction techniques can be obtained [17]. A good understanding of the antenna and its patterns was obtained from the official documentation made by Cisco [18]. Outdoor radio wave propagation of 3D models can be obtained by using assumptions that could simulate parameters to real measurements [18].
Wireless InSite will be used to provide propagation modeling for the two protocols. A white paper written by Remcom, gives some new supportive insight into the ray tracing code by giving a specific example of RF modeling of mines [19]. There is variation in the throughput produced for LOS and NLOS of 802.11n using MIMO antennas in an outdoor environment. This investigation found a correlation in the 2.4 GHz and 5 GHz [20].
EDX Signal Pro is a general purpose software package which provides a complete set of planning tools for wireless networks from 30 GHz to 60 GHz [21]. It has full access to the terrain along with a demographic coverage to provide point to point analysis. Belu et al attempt to develop a MATLAB application that creates RF propagation models, along with a user interface for educational purposes [22][23].
While a lot of work has been already done on building transceivers for 802.11n and LTE, none of them have been specifically designed for operation in space. Furthermore, until now LTE and 802.11 protocols have not been tested against each other .We built on this existing research by implementing both LTE and 802.11n on an SDR and testing them for data rate, throughput and multipath fading around the ISS.
III. Research Methodology
Sub-problem 1: Selection of RF Modeling Tools
A. Selection of Software Tool for Propagation Modeling
RF modeling of the ISS infrastructure requires a software that allows the placement of both RF transmitters and receivers that would in turn generate a specified output. Software such as MATLAB, EDX Signal Pro and Ekahau did not support this capability.
Although MATLAB has a dedicated WLAN and LTE toolbox, the latest version does not possess extensive RF propagation modeling functionality. MATLAB does not allow the placement on antennas/ access points on a 3D model along with no functionality to generate radiation patterns. Detailed heat maps demonstrating coverage also could not be generated and therefore this software was not used.
Ekahau can create heatmaps for numerous parameters such as received power, AP coverage and throughput. However, it lacks the ability to import three dimensional structures, such as the ISS, for propagation modeling.
EDX Signal Pro was a very efficient signal modeling tool but the only setback, like Ekahau was that it did not allow modeling of 3D objects.To potentially utilize the EDX tool, it is required to have a 3D floor plan, which was not made available by NASA for public use.
For all the above reasons, Wireless InSite was chosen as the modelling platform.
B. Selection of Software and Hardware Platform for SDR Design
Several vendors market various SDRs with different capabilities, and as such it becomes important to select a SDR that supports the requirements of the project. The following points will describe the research methodology for selecting an appropriate SDR for the research project:
a) Modulation Scheme
As a first step, 802.11n and LTE wireless protocols were evaluated. The 802.11n wireless protocol has various modulation schemes for wireless communication. Depending on the spatial streams, different modulation techniques can be used. This project evaluates 2×2 spatial stream systems that support BPSK, QPSK, 16-QAM and 64 QAM modulation techniques. Thus, a software defined radio hardware that supports the above mentioned or newer modulation schemes such as 256 QAM would be the ideal hardware module to run Wi-Fi and LTE.
b) Hardware Compatibility
SDRs have different hardware formats such as USRP and WARP. SDRs can also be categorized by vendor. Each vendor supports different frequency ranges of operation, sampling rate, data rate, bandwidth and host interface. Ideally, the solution must use an SDR that supports both LTE and 802.11n protocol specific characteristics. SDRs that work across multiple operating systems are advantageous as this feature increases its usability. Many SDR platforms support open source software, making Linux a strong candidate operating system. Also, an appropriate vendor specific SDR has to be selected that supports the working of both 802.11n and LTE wireless protocol that is compatible to be used on all three operating systems mentioned above. A detailed market research of available hardware platforms was carried out as shown in Table 1.
A Universal Serial Radio Peripheral (USRP) based SDR was used to generate waveforms for wireless communication. A USRP based hardware platform is comparatively inexpensive and has good performance. This hardware platform can then be used in conjunction with a host computer. The host computer was connected to the USRP using a high-speed link. The host based computer was used to control the USRP using host based software (GNU or MATLAB) [34]. The USRP consists of a motherboard that has a clock, for synchronization of data, FPGA, ADCs, processor interface and power that can be regulated. These components are used for baseband processing of signals. The USRP is normally controlled by an open source UHD driver. The UHD driver supports Linux, MacOS, and Windows platform. It also supports several software based platforms like GNURadio, LabVIEW, MATLAB and Simulink.
c) SDR Design
For designing the SDR to emit radio frequency waves as per the LTE or 802.11n wireless protocols, software tools must be used. Some of the software specific tools that support SDR design modeling are GNURadio, MATLAB and LabVIEW. Each of these software design tools work differently for different SDRs. To form the base of a solid recommendation as to which software tool could be ideal, an extensive market research was carried out to compare several open source and pay-for-license software platforms as shown in Table 2.
Company
Product
Tx/Rx
Freq. MHz
BW MHz
List Price
Web
Airspy
Airspy R2
Rx
24-1800
10
$199
airspy.com
Airspay mini
Rx
24-1800
6
$99
Ettus Research
Universal Software Radio Peripheral (USRP) Series
www.ettus.com
B200
Tx/Rx
70-6000
56
$686
B210
Tx/Rx
70-6000
56
$1,119
USRP1
Tx/Rx
$719
N200
Tx/Rx
0-6000
40
$1,541
N210
Tx/Rx
0-6000
40
$1,746
E310
Tx/Rx
70-6000
56
$2,746
E312
Tx/Rx
70-6000
56
$2,990
E313
Tx/Rx
0-6000
56
$3,750
X300
Tx/Rx
0-6000
160
$3956
X310
Tx/Rx
0-6000
160
$4871
Nuand
BladeRF x40
TX/RX
300-3800
28
$420
www.nuand.com
BladeRF x115
TX/RX
300-3800
28
$650
RTL-SDR
DVB-T DAB FM SDR
Rx
24-1766
3
$20
rtl-sdr.com
Table 1 – Hardware platforms for SDR
Platform
Module
Cost
Code Maturity
Community Involvement
Hyperlink
GNU Radio
gr-lte
Open Source
Expired
Support Thread: Inactive
https://github.com/kit-cel/gr-lte
gr-IEEE 802.11 a/g/p
Open Source
Mature
Support Thread: Active
https://github.com/bastibl/gr-ieee802-11
MATLAB
Mathworks LTE SIB1 2 Channel Tx Rx
Open Source
Mature
Support Thread: Active
https://www.mathworks.com/help/lte/examples/two-antenna-lte-sib1-transmission-using-usrp-based-software-defined-radio-sdr.html
Mathworks 802.11 OFDM Beacon Receiver
Open Source
Mature
Support Thread: Active
http://www.mathworks.com/help/supportpkg/usrpradio/examples/ieee-802-11-tm-wlan-ofdm-beacon-receiver-with-usrp-r-hardware.html
Virtual Network Communications
Micro EPC
$8000
Commercially Available
Tech Support Available
http://www.virtualnetcom.com/ProductsMicroEPC.html
Hosted LTE core
$6500
Commercially Available
Tech Support Available
http://www.virtualnetcom.com/ProductsHostedLTE.html
GreenCell
$5000
Commercially Available
Tech Support Available
http://www.virtualnetcom.com/ProductsGreenCell.html
Table 2- Software platforms for SDR
Sub-problem 2: Comparison of LTE and Wi-Fi
Despite having numerous disadvantages in terrestrial environments, Wi-Fi currently is still more popular than LTE networks, mostly due to reduced cost, perceived reliability and speed. It is interesting to note, however, that LTE is less affected by multipath channel and significant levels of Doppler Shift and provides better coverage as compared to 802.11 Wi-Fi networks. Space networks have a different set of requirements that a wireless technology must satisfy including support for multi hop relay communications, higher data rates to collect high volumes of data from sensors and international protocol support.
The research makes a theoretical comparison between the two protocols based on the following parameters:
1. Comparing the specifications of 802.11n, 802.11ac and LTE-Advanced networks.
2. The peak uplink and downlink data rates of each protocol for a 4×4:4 system, i.e. a system that has 4 transmitters, 4 receivers and is capable of transmitting 4 simultaneous spatial streams.
3. The maximum range demonstrated by each of these wireless protocols.
4. The spectral efficiency (bits per seconds/Hertz) of the protocols
For a 4×4:4 system, LTE, 802.11n and 802.11ac can be compared as shown in Table 3.
Wi-Fi 802.11n
Wi-Fi 802.11ac
LTE-A
Emphasis
Speed over short distances
Speed over short distances
Range and mobility, lesser data rates
Architecture
Simple
Simple
Complex
Peak Uplink
600 Mbps
866 Mbps
3 Gbps
Peak Downlink
600 Mbps
866 Mbps
1.5 Gbps
Max. range
32m
50 m
Up to 100 km.
Spectral Efficiency
(per spatial stream)
SISO: 3.75 bps/Hz
SISO: 5.5 bps/Hz
SISO: up to 30 bps/Hz
Table 3 – Comparison of wireless protocols
Sub-problem 3: RF Propagation Modeling
A. How does Wireless InSite serve the purpose?
RF modeling is performed on a 3D object file of the International Space Station that is in .stl format. This file is imported in the software, scaled to real world dimensions, after which the material of the object, in this case metal, is specified to model accurate multipath environment and attenuation characteristics. A study area that fits the entire dimension of the object is selected. The faces of the object are selected to place transmitters and receivers. Specific parameters such as frequency, antenna type and even the transmit power (Tx) are set. InSite also allows the user to build MIMO antennas. For some of the models generated for this project, a 2×2 MIMO antenna with constituent half wave dipole antennas were utilized.
After the antennas have been placed, a simulation of the project is run. This run creates a calculation window which shows the time span to compute propagation paths of each transmitter. A specific function called the communication system analysis shows the path, power of each transmitter (Tx) and receiver (Rx). This can be easily loaded and viewed on the 3D object or can be plotted on graphs usually against distance or time. The plot obtained on the object has a legend below it that is made up of various colors which helps to understand the different numerical parameters in each region. The heat map, in the form of an XY Grid is made after placing receiver points on the object. This depicts the coverage of the placed antennas on the object, and the extent to which each region receives the signal. The project involves a baseline model which uses the current EHDCA architecture, with four transmitters on the side and a receiver in the center of the ISS. Another improvised model is composed of twelve transmitters placed in various optimized positions of the model along with seven receivers to establish stronger coverage capability. The improvised model is used to test LTE and 802.11 protocol by changing the frequency and running it with and without MIMO capability. The figures and resulting parameters of the simulation can be viewed in the Results section.
B. Establishing the feasibility of Wi-Fi and LTE on SDR
The RF modeling carried out using Wireless InSite demonstrates the capabilities of Wi-Fi and LTE in the form of heat maps around the 3D model of the ISS. To back the recommendation of a wireless protocol that should be deployed external to the ISS, Wi-Fi and LTE were both implemented on the Ettus Research B210 Radio in order to establish that the B210 radio is capable of handling the processing power required to run these protocols. To set a baseline, only signaling messages were transmitted and received. The design codes used for this purpose are – “LTE SIB1 Transmission over Two Antennas” and “IEEE 802.11 WLAN – Beacon Frame Receiver of USRP Hardware”. Both these codes are a part of the list of examples developed by Mathworks for use on MATLAB. The result of this experiment was a successfully designed radio pair which could transmit and receive SIB1 messages (in the case of LTE) and OFDM beacons (in the case of Wi-Fi). The resulting graphs are included in the results section.
IV. Results
I. Wireless InSite
The first model is the EHDCA architecture which is currently deployed external to the ISS. This is the baseline model against which all the other models are compared. As seen in the Fig. 3, it has four transmitters and one receiver in the center. The throughput obtained is as low as 34 Mbps on the entire ISS, with 30 Mbps on the left-hand edge. The following results will show how the coverage can be improved by using Wi-Fi or LTE along with an increased number of transmitters and receivers.
The Baseline EHDCA Architecture (Fig. 3) has a throughput of 34Mbps while the maximum throughput covers 90% of the area.
Fig. 3 – Baseline EHDCA Architecture – 802.11n- Received Power
In the model shown in Fig.4, Wi-Fi (5GHz, 40MHz bandwidth) with MIMO has been implemented on twelve transmitters and seven receivers. Each Tx and Rx are 2×2 MIMO antennas, and Tx power is set to 30dBm.
The Wi-Fi with MIMO (Fig. 4) has a throughput of 34Mbps while the maximum throughput covers 94% of the area.
Fig. 4 – Wi-Fi with MIMO- Received Power
The same antenna placement is used for LTE with MIMO i.e Fig 5 and Fig 6. The antenna type chosen is MIMO, while the TX power is kept to 30dBm.
The LTE with MIMO (Fig. 5 And Fig. 6) has a mode throughput of 65Mbps, maximum throughput of 75Mbps and lowest throughput of 51Mbps. The combined coverage provided by the mode and maximum throughputs is 70% of the area.
Fig.5- LTE with MIMO- Received Power
Fig.6 – LTE with MIMO- Throughput
In case of Wi-Fi without MIMO (Fig. 7 and Fig. 8), while the number, placement and transmit power (30dBm) of antennas are unchanged, the type of antenna selected is half wave dipole.
The modal throughput is 34Mbps while the maximum throughput covers 96% of the area.
Fig.7- Wi-Fi without MIMO- Received Power
Fig.8 – Wi-Fi without MIMO- Throughput
Similar to the Wi-Fi without MIMO model, the LTE without MIMO models (Fig. 9 and 10) have been simulated with 12 Transmitters and 12 Receivers, with each transmitter having an input power of 30dBm. The antenna type chosen is Half-wave Dipole.
The modal throughput of 65Mbps, maximum throughput of 75Mbps and lowest throughput of 51Mbps. The combined coverage provided by the mode and maximum throughputs is 80% of the area.
Fig.9 – LTE without MIMO-Received Power
Fig.10 – LTE without MIMO- Throughput
The final results for the Wireless InSite simulation of the LTE and Wi-Fi models can be summarized in Table 4.
Baseline EHDCA Architecture
Wi-Fi with
MIMO
Wi-Fi without MIMO
LTE with
MIMO
LTE without MIMO
Received Power Range
-25.9 dBm to -77.5 dBm
-19.9 dBm to
-82.8dBm
-16.9 dBm to
-75.9
dBm
-4.8 dBm to – 61.3
dBm
-1.9 dBm to -61 dBm
Throughput Range
30 Mbps to
34 Mbps
30 Mbps to 34Mbps
15 Mbps to 34 Mbps
51 Mbps to 75 Mbps
51 Mbps to 75 Mbps
Mode Throughput
34 Mbps
34 Mbps
34 Mbps
65 Mbps
65 Mbps
Coverage Area – Mode Throughput
90%
94%
96%
60%
50%
Coverage Area – Max Throughput
90%
94%
96%
10%
30%
Table 4 – Comparison table for propagation models
This table describes the measured parameters for each of the propagation models simulated on Wireless InSite. While both Wi-Fi and LTE have 100% coverage for the study area, the lowest throughput measured for Wi-Fi models is 34 Mbps and for LTE the lowest throughput measured is 51 Mbps. This demonstrates LTE to have better coverage and capacity.
II. SDR Design
A. LTE on SDR:
Software Platform: MATLAB
Hardware Platform: Ettus Research USRP B210
Test Setup: The Experimental setup for transmitting LTE SIB data is shown the Figure 11.
Fig. 11 – LTE on SDR Tx/Rx
MATLAB R2016a version was used to set up this test environment along with the Ettus research B210 software defined radios. Two laptops, one running the LTE transmitter code and the other running the LTE receiver code were used simultaneously for transmission and reception of LTE based SIB data frames. The Transmitting SDR had a red led glowing while the data packets were being sent to the receiving SDR. Also, upon reception of data frames from the transmitting SDR, the Receiving SDR had a green LED glow to show the successful transmission of data.
Figure 12 shows the frequency spectrum that illustrates the fact that there were two channels on which the LTE SIB data was transmitted. On the receiver end, the frequency spectrum depicted in figure 13 shows the receiving channel spectrum. It is evident from the figure that during reception of data, there was significant fading in signal. This was due to the fact the test was conducted in a very noisy environment with lot of interference nearby. The center frequency used for this test was 900 MHz at a sampling rate of 19200 samples per frame. There were 4 frames in one signal burst of sample. The master clock rate was set at 7680000. The modulation scheme used for the transmission of data was QPSK. Even though there was significant signal fading observed through the received signal spectrum, there was no observed resampling of signal frames.
Fig.12. Transmitter Spectrum
Fig 13. Received spectrum
Figure 14 suggests that there were a lot of symbols that were scattered which shows that the signal received at the received end was considerably distorted and of inferior quality.
Fig. 14. Constellation diagram.
B. Wi-Fi on SDR
Software Platform: MATLAB
Hardware Platform: Ettus Research USRP B210
The experimental setup for this test also included MATLAB R2016a version and Ettus research B210 software defined radio. This test consists of the usage of only one laptop that decodes beacon frames the transmitting beacon frames from nearby access points. The received code was locked at the 5GHz frequency range to listen for beacon frames. As observed in the MATLAB results, beacon frames from the SSIDs “Dash Wireless 5G”, “UCB Guest” and “eduroam” were processed after an interval of 100 and 200 milliseconds each. One packet was decoded from Dash Wireless 5G, three packets from UCB Guest and four packets from eudroam SSID. These beacon frames were received and decoded on channels 153 and 157 of the 5 GHz frequency band.
This experiment successfully demonstrates the feasibility of Wi-Fi on SDR.
V. Conclusion
The theoretical analysis as well as Wireless InSite Simulations clearly demonstrate that LTE is a better protocol for capacity requirements of the ISS. The maximum and mode throughput obtained for LTE spikes up to twice the corresponding throughput values measured in the Wi-Fi model. Although the coverage for the Wi-Fi without MIMO model is 96%, the mode throughput is only 34Mbps. The LTE without MIMO model, on the other hand, has a coverage of 50% with a throughput of 65Mbps. This LTE model serves higher capacity requirements as 80% of the coverage area has a throughput of 65Mbps or more. For both Wi-Fi and LTE, the models with the half-wave dipole antenna performed better than models with MIMO antennas.
The market research carried out in the due course of this research project recommends that the Ettus Research B210 hardware along with MATLAB is one of the best possible ways to deploy Wi-Fi and LTE on a SDR. MATLAB can also be substituted with other paid software platforms such as Open Air Interface, which is a renowned module for LTE design.
This research could further lead to investigations on transmit power limits and policies for wireless communications that could be set for space and its external environment. It could also help NASA in setting up various Wi-Fi networking models and in managing SDR deployments. Also, as a future task, atmospheric conditions in the space environment could be incorporated into RF coverage planning and SDR design. Appropriate hardware can be selected based on these results to withstand the extreme conditions in space. This would render a communications model appropriate for deployment in space.