With the technological challenges surrounding Software Defined Radio (SDR) largely overcome, the deployment of first generation sets is underway. Possible enhancements such as increasing the bandwidth while simultaneously reducing the size, weight, power, and cost of the device have yet to be developed. Beyond this, Cognitive Radio represents the next major advance in wireless technology, holding out the tantalizing promise of significantly greater spectrum utilization, however, there are some significant barriers which still must be surmounted.
Software Defined Radio (SDR)
Software Defined Radio has evolved over the last decade to become the latest generation of wireless technology for the modern war fighter. This has been spurred by innovations in programmable hardware that have allowed unprecedented sophistication in software. Hardware advances in both the logic and RF domains have enabled the creation of radios with far greater capabilities in much smaller form factors. For example, the capacities and additional features built into FPGAs have grown exponentially. Today it is possible to design using a 20nm device from Intel (formerly Altera) boasting up to 660k logic elements, 39 Mbits of on-chip memory, over 3,300 DSP blocks, 48 serial transceivers each capable of 17.4 Gbps, and dual 32-bit ARM processor cores clocking at 1.5 GHz. These offer baseband processing far in excess of what was feasible even a few years ago.
Similarly, Field Programmable RF (FPRF) devices have become far more capable. The first generation FPRF from Lime Microsystems broke all the barriers for integration when it was launched. It offers a complete transceiver integrating RF amplifiers, filters, mixers, and gain blocks, along with Low Noise Amplifiers (LNAs) and data converters. The resultant single chip enables highly capable SDR sets to be produced with superior size, weight, power, and cost, commonly abbreviated to SWaP-C.
Today, by combining a second-generation low power FPRF (LMS7002M) with the newly-released Up/Down RF frequency shifter devices, designers have unsurpassed RF capabilities at their disposal. The FPRF device features dual transceivers and the chip has field programmable DSP blocks and Received Signal Strength Indicators (RSSIs) to bolster the performance while at the same time reducing the power consumption. FPRFs intrinsically cover the frequency range 100 kHz to 3.8 GHz, which encompasses a multitude of bands used by air, sea, and ground forces. The ability, in a single chip, to support such a range of frequencies makes it a very capable device for SDR applications.
LMS7002M Dual transceiver FPRF block diagram (Source: Paul Dillien)
The FPRF also supports a wide range of modulation modes, such as Multiple-Input Multiple-Output (MIMO) and all the variations of Code Division Multiple Access (CDMA). The field programmable feature allows the device to be very rapidly retuned, thereby providing unprecedented agility for frequency hopping and other interception counter-measures.
The companion chip to the FPRF is a quad Up/Down RF frequency shifter called the LMS8001 from Lime Microsystems. This extends the range of the LMS7002M FPRF from 100 kHz to 12 GHz greatly adding to the SDR possibilities. The fully programmable frequency shifter features four wide-band LNAs, RF mixers, and power amplifier drivers along with a local oscillator.
LMS8001 Quad up/down frequency converter block diagram (Source: Paul Dillien)
Signals up to 12 GHz are converted up or down into the range that the FPRF can process. This capability offers some powerful options. For example, the two transceiver channels can be tuned to different bands to allow channel aggregation that doubles the 80 MHz digital bandwidth to 160 MHz. This is achieved by using the up/down converter to introduce an offset in frequency into the receiver and transmitter paths of one FPRF channel. This uses just two of the available four channels in the converter.
The digital baseband chip forms a key part of the SDR system as it processes the signals to be transmitted and decodes the received data patterns. The modulating signal is first expressed as the amplitudes of the in-phase and quadrature components relative to the carrier phase. The encoding and decoding algorithms run in the FPGA programmable fabric. Similarly, data encryption can take advantage of the high speed parallel processing available in hardware.
The SDR needs to monitor the spectrum for other traffic, including attempts to jam the signal. The FPRF provides an RSSI for both the RF and the baseband signal levels. The RF level detector can be used as part of the system automatic gain control and will indicate when interfering or jamming signals are present.
Data bandwidth is a vital part of a modern ISTAR (information, surveillance, target acquisition, and reconnaissance) system. The bandwidth and resilience of a wireless system can be enhanced with the latest modulation techniques such as spatial MIMO. This is a scheme that uses two or more antennas and improves the spectral efficiency while also achieving a diversity gain that improves the link reliability. The 2×2 MIMO can be fully supported in a single FPRF device, while higher levels such as 4×4 can be realized with two FPRF chips. MIMO encoding, and especially the complex decoding, require more capable baseband chips, which are now available in the form of the latest FPGAs.
War fighters may find they are out of range or jammed from communicating with terrestrial-based infrastructure. SDR might be able to circumvent jamming by switching or hopping to an alternative frequency that is not jammed. The FPRF, as the name implies, is also capable of being reprogrammed in the field, and this feature can be exploited to rapidly change the frequency band of the transmitter or receiver, thereby allowing it to hop to another band to avoid detection or jamming.
Once more, advanced air interfaces can also help. CDMA technologies allow users to communicate via voice, data, and video simultaneously at all levels of security. CDMA uses multiple carriers that are spread across a wide frequency band to carry the data. This technique allows the signal level of each carrier to be much closer to the ambient noise level, and the transmission is therefore more difficult for an adversary to detect and intercept. Coding schemes include forward error correction, which uses redundancy added into the encoding to remove errors at the receiving end. This gives robustness against noise and transmission errors caused by fading, and it also makes it harder for an enemy to jam the signals.
When the user is out of range in a mountainous region, a different solution is required. Equipment that can connect to a satellite, such as an X-band transceiver, solves this issue and is less affected by rain fade than Ka band links. The X band uses 7.9 GHz to 8.4 GHz as an uplink, and frequencies in the range 7.25 GHz to 7.75 GHz for downlink communication. These are within the range now available from FPRF-based systems featuring up/down converters. Moreover, it is possible to build a compact, highly-integrated RF system that combines both satellite and terrestrial links, because the devices can be configured to simultaneously support both links.
Dual-mode (terrestrial/satellite) SDR concept (Source: Paul Dillien)
The use of a dual transceiver FPRF and two RF frequency shifters allows one channel to be used for the satellite link and the second for ground communications. It is also possible with this configuration to act as a repeater station that relays the signals between the two transmission systems.
Next-generation radio sets
The next evolution beyond SDR is expected to be the design and deployment of Cognitive Radio (CR). This sophisticated scheme builds on an SDR capability, but has some significant differences. Rather than the operator selecting the band and frequency to use, CR systems negotiate to exploit free channels and adapt to the current environment. This can significantly increase the utilization of the available spectrum, which is a heavily used resource during battle.
The CR must scan the bands of interest before setting up a call to establish what resources are free. Scanning will also identify jamming signals and will mark the frequencies as occupied. This is done by monitoring the outputs of the on-chip RSSIs while the FPRF is programmed to ramp through the frequency band. RSSIs are provided in both the RF and base band domains, so that activity and signal strength can be readily calculated.
Cognitive Radio (CR) concept (Source: Paul Dillien)
CR-enabled radios need to establish and agree a communication channel to set up a call with another set. They must have a method of negotiating what frequency, power levels, and modulation scheme to adopt. One option is to use a base station setup at the Command Post (CP) to pass messages between the participants. The CP contacts the called CR on a dedicated channel and establishes an information exchange of connection data, and then the CR radios switch to the agreed settings. A further option is to use a satellite link back to the CP to agree the parameters of the link, and to then free up the satellite for other traffic.
During the transmissions, the CR equipment needs to continuously monitor the channel and other frequencies, because other (non-CR) users operating legacy equipment may start up a call. In this eventuality, the CR role will be to sense the new transmission, establish and agree another free frequency, and disconnect from the current settings. To facilitate this, the CR spectrum-sensing methodology needs to provide time slots when it is not transmitting. This is to enable the system to detect other signals and it can be accommodated within the frame format for the overall CR system. The CP can collate reports on spectral usage from several CR sets to update a table of free and busy channels. The CP can use this table to allocate free channels against requests for service. The algorithm can protect against excessive hunting by a CR when a new network configuration is required. This cooperative scheme also mitigates the “hidden node” problem whereby a transmission from a remote or shaded user is unnoticed by a particular CR. The direct signal may be too weak to be detected, but it would cause interference to the existing user if the CR started to transmit at the same frequency.
Cognitive Radio detecting other traffic and switching to a free channel (Source: Paul Dillien)
The operation of a CR in a typical wireless environment requires constant monitoring. Initially, the CR detects that Frequency C is busy and that Frequencies A and B are currently unused with just wide-band noise detected. The CR systems communicate via the CP and agree to use Frequency A. During the CR call, one of the CR sets detects that a non-CR user has connected or a jamming signal starts up, and so the CR callers evaluate other options and switch to Frequency B that is still only registering noise. Once again, during the CR call, new traffic is detected, and so the CR call switches to Frequency C, which is now unused.
Issues to full deployment
Overcoming the technical obstacles is only part of the problem. Before CR can become a truly mainstream military wireless technology, there needs to be agreement between the different branches of the armed forces that spectrum sharing will benefit everyone. Large swaths of valuable spectrum could be opened up to CR users given sufficient assurances on interference mitigation.
Cognitive Radio is currently the subject of initial research, but it promises to provide a highly flexible and robust system that eases the problem of spectrum scarcity. The latest innovations in programmable wireless and programmable logic bring us a step closer to realizing this potential.
Paul Dillien is the founder of High Tech Marketing. He is a Chartered Engineer and MIET who has worked in strategic and tactical marketing roles and has specializations in competitive analysis and negotiation. Paul has over 30 years' experience working for UK and US companies in the semiconductor industry. He is also the author of The FPGA Market Report , which provides insights into this complex market (click here for more details).