Abhishek Ambede

A modified coefficient decimation method to realize low complexity FIR filters with enhanced frequency response flexibility and passband resolution

Low complexity, reconfigurable finite impulse response (FIR) filters using coefficient decimation method (CDM) has been recently proposed in literature. In this paper, we propose a modified coefficient decimation method (MCDM) which enhances the flexibility of CDM in obtaining FIR filters with varied passband locations. The resolution of the center frequency locations in the multi-band frequency responses obtained using MCDM is twice that of the conventional CDM. Further, the stopband attenuation of FIR filters realized using our MCDM is higher compared to the filters obtained using conventional CDM. Also, for the same prototype modal (original) filter, the number of distinct frequency band locations that can be obtained after coefficient decimation is greater for MCDM than CDM due to the increased center frequency resolution of the former method. The hardware realization architecture of MCDM is presented. The advantages of our method for realizing reconfigurable filter banks are also discussed.

Flexible Low Complexity Uniform and Nonuniform Digital Filter Banks With High Frequency Resolution for Multistandard Radios

Multistandard radios typically employ digital filter banks (FBs) for channelization of wideband input signals consisting of multiple radio channels corresponding to different wireless communication standards. In this paper, we propose a new design technique to obtain uniform as well as nonuniform digital FBs for multistandard channelization. In the proposed FB design technique, the improved coefficient decimation method (ICDM) is used to obtain different low-pass, high-pass, and multiband frequency responses using a single low-pass prototype filter. These frequency responses are algebraically operated upon using appropriate spectral subtraction, complementary filter response operation, and frequency response masking operations to obtain the desired sub-bands in the FB. With the help of suitable design examples, we show that our FB is a low complexity and flexible alternative to the other digital FBs in the literature, in both uniform and nonuniform channelization scenarios. The proposed ICDM-based FB is shown to be able to achieve a higher number of distinctly located sub-bands with twice the center frequency resolution when compared with the discrete Fourier transform-based FB (DFTFB) and the conventional coefficient decimation method-based FB (CDFB). For a uniform channelization design example for wireless communications, implementation results show that the proposed ICDM-based FB achieves 70.92% and 15.49% reductions in resource utilizations and 58.72% and 23.13% reductions in power consumptions when compared with the DFTFB and CDFB respectively. Also, for a nonuniform channelization design example for wireless communications, the proposed ICDM-based FB shows an 11.46% reduction in multiplication complexity when compared with the CDFB.

Design and Realization of Variable Digital Filters for Software-Defined Radio Channelizers Using an Improved Coefficient Decimation Method

Variable digital filters (VDFs) are used in software-defined radio handsets for extraction of individual radio channels corresponding to multiple wireless communication standards. In this brief, we propose a VDF based on the improved coefficient decimation method. The proposed VDF provides variable low-pass, high-pass, bandpass, bandstop, and multiband frequency responses on the fly, using the same set of prototype filter coefficients. We present nonpipelined and pipelined implementation architectures for the proposed VDF, along with field-programmable gate array implementation results for multiple VDF designs. Analysis of the implementation results shows that the pipelined implementations achieve average reductions of 27.66%, 49.17%, and 25.59% in the number of occupied slices, dynamic power, and energy consumption, respectively, when compared with corresponding nonpipelined implementations. In addition, the proposed pipelined implementation architecture provides high operating frequencies that are independent of the prototype filter order across different VDF designs. An average maximum frequency of 157.89 MHz is obtained.

An improved coefficient decimation based reconfigurable low complexity FIR channel filter for cognitive radios

Multi-standard channel adaptation is a critical function in cognitive radio handsets which involves the transmission/reception of individual frequency channels of multiple wireless communication standards at different intervals of time. This needs dynamically reconfigurable, low complexity and high speed digital channel filters. In this paper, we present a reconfigurable finite impulse response (FIR) channel filter design technique based on the combination of the conventional coefficient decimation method (CDM) and a modified CDM. Our method enhances the frequency response flexibility of the filter and doubles the center frequency resolution when compared to the conventional CDM. The proposed channel filter has a significantly lower multiplication complexity and achieves superior stopband and transition band characteristics when compared to the channel filters based on the conventional CDM. Design example shows that a 57.1% reduction in multiplication complexity is achieved if the proposed channel filter is designed instead of the conventional CDM based channel filter.

Design of a low complexity channel filter satisfying LDACS1 spectral mask specifications for air-to-ground communication

Modernization of the existing air traffic management systems is currently under way to support the sustained air traffic growth worldwide. A new standard called L-band Digital Aeronautical Communication System (LDACS) is being developed for air-ground communications, which aims to facilitate improved aeronautical communications in the next generation air traffic management systems. LDACS type 1 (LDACS1), a broadband system based on orthogonal frequency-division multiplexing (OFDM) modulation, is the most promising and mature candidate for LDACS. The channel filter in a LDACS1 receiver must be designed with a stringent spectral mask to avoid interference from the adjacent Distance Measuring Equipment (DME) channels as well as the neighboring LDACS1 channels. In this paper, we present the design of a low complexity channel filter for LDACS1. Design of the proposed channel filter is based on the frequency response masking technique. When compared with the conventional channel filter design approach, the proposed channel filter offers 74.26% savings in multiplication complexity and 17% reduction in group delay.

Design and Implementation of High-Speed All-Pass Transformation-Based Variable Digital Filters by Breaking the Dependence of Operating Frequency on Filter Order

All-pass transformation (APT)-based variable digital filters (VDFs), also known as frequency warped VDFs, are typically used in various audio signal-processing applications. In an APT-based VDF, all-pass filter structures of appropriate order are used to replace the delay elements in a prototype filter structure. The resultant filter can provide variable frequency responses with unabridged control over cutoff frequencies on the fly, without updating the filter coefficients. In this brief, we briefly review the first- and second-order APT-based VDFs along with their hardware implementation architectures, and provide generalized design procedures to realize them as per required specifications. We also propose the modified pipelined hardware implementation architectures for both the first- and second-order APT-based VDFs. Field-programmable gate array implementation results of different first- and second-order APT-based VDF designs for both nonpipelined and pipelined implementation architectures are presented. An analysis of the results shows that the proposed pipelined implementation architectures result in high-speed VDFs, achieving high operating frequencies that are independent of the prototype filter order, for both the first- and second-order APT-based VDF designs.

Efficient FPGA implementation of a variable digital filter based spectrum sensing scheme for cognitive IoT systems

Wireless communication in Internet of Things (IoT) systems involves several devices communicating in the limited available frequency spectrum using different wireless communication standards. To achieve efficient spectrum usage in the IoT systems, it is desired to have the ability of dynamic spectrum allocation. Cognitive radios (CR) enable efficient spectrum utilization by performing spectrum sensing to gain the knowledge of current spectrum occupancy and dynamically reconfiguring the communication parameters, making them highly suitable for IoT systems. In this paper, we present the efficient hardware implementation of a spectrum sensing scheme for CR based IoT applications that require large bandwidth. We show that the proposed implementation on field programmable gate arrays (FPGAs) can achieve over 1.56x improvement in performance, while consuming 77% less generic resources and at a fractionally higher power consumption.

Optimisations in aeronautical communications using aircrafts as relays

The airline industry has seen a tremendous growth in the global air traffic over the past few years and studies forecast this rapid growth to sustain in the future as well. This has led to multiple challenges for the smooth operation and management of air traffic all around the world. For instance, the currently deployed air-ground communication systems are forecast to reach their capacity limits in the next few years. The International Civil Aviation Organization has thereby initiated numerous projects to develop new technologies for the next-generation of air-ground communication systems. In this paper, we investigate use of relaying technique for air-ground communication wherein intermediate aircrafts are used to relay (receive and forward) the communication signals from other aircrafts to the ground stations. We aim to exploit the strengths of the air-air communication channels to aid the communication in the air-ground channels. We propose the usage of aircrafts which are nearer to the ground station for relaying messages of aircrafts located further away to improve the quality of communication. Our simulation results show that this can enable higher data rates and reduce the overall power consumption when compared with the direct air-ground communication technique that is currently used.

Design of a low complexity filter bank satisfying LDACS1 spectral mask specifications for base-station receivers in air-ground communications

The sustained air traffic growth worldwide has resulted in the need for upgrading the existing global air traffic management (ATM) system. Consequently, a new standard called L-band Digital Aeronautical Communication System (LDACS) is being developed for the air-ground communications component of the next generation ATM systems. Amongst the options being considered for LDACS, the LDACS type 1 (LDACS1) is the most superior and mature candidate and is likely to be the final choice for deployment. The presence of previously deployed and operational legacy systems in the L-band has led to stringent spectral mask specifications for LDACS1, to ensure there is no interference caused to the former. In this paper, we propose the design of a digital filter bank which can satisfy the LDACS1 spectral mask specifications while also achieving low complexity of implementation. Design of the proposed filter bank is based on the improved coefficient decimation method. With the help of a design example, we show that when compared with the discrete Fourier transform based filter bank, the proposed filter bank achieves 71.49% reduction in multiplication complexity.

A power and time efficient radio architecture for LDACS1 air-to-ground communication

L-band Digital Aeronautical Communication System (LDACS) is an emerging standard that aims at enhancing air traffic management by transitioning the traditional analog aeronautical communication systems to the superior and highly efficient digital domain. The standard places stringent requirements on the communication channels to allow them to coexist with critical L-band systems, requiring complex processing and filters in baseband. Approaches based on cognitive radio are also proposed since this allows tremendous increase in communication capacity and spectral efficiency. This requires high computational capability in airborne vehicles that can perform the complex filtering and masking, along with tasks associated with cognitive radio systems like spectrum sensing and baseband adaptation, while consuming very less power. This paper proposes a radio architecture based on new generation FPGAs that offers advanced capabilities like partial reconfiguration. The proposed architecture allows non-concurrent baseband modules to be dynamically loaded only when they are required, resulting in improved energy efficiency, without sacrificing performance. We evaluate the case of non-concurrent spectrum sensing logic and transmission filters on our cognitive radio platform based on Xilinx Zynq, and show that our approach results in 28.3% reduction in DSP utilisation leading to lower energy consumption at run-time.