Power and sub-channel optimization of JPEG 2000 image transmission over OFDM-based cognitive radio networks

https://doi.org/10.1016/j.image.2017.08.002Get rights and content

Highlights

  • A new unequal power and channel allocation algorithm is proposed for transmission of JPEG 2000 images over cognitive radio systems.

  • The proposed algorithm enhances the average PSNR significantly for multipath channels.

  • The algorithm can be implemented in real time.

  • The scheme can be generalized to other standards such as JPWL

Abstract

Cognitive Radio is an efficient way of spectrum utilization since secondary users with bandwidth-demanding applications such as multimedia can get access to licensed frequency resources opportunistically and resolve their bandwidth limitations. Among all multimedia formats, JPEG 2000 is a suitable candidate for cognitive radio networks thanks to its unique features. In conventional resource allocations for Cognitive Radio systems, all data bits are assumed equally important. However, different parts of the JPEG 2000 bit stream have different contributions to the quality of the received image. Therefore, in this paper, an unequal power allocation method is used to allocate the available power to the coded bits based on their importance in the image quality. Simulated annealing method is used to solve the unequal power allocation problem. Furthermore, bits with higher significance are further protected by using sub-channels with better channel quality. Thus, the likelihood of significant bits being received correctly is increased. The optimal solution is obtained by minimizing the image distortion without violating the interference constraint to the primary users. The performance of the proposed method is demonstrated by simulation results.

Introduction

The rapid developments in technology have made all types of applications available in a single mobile handset, and multimedia applications are the most popular. The growing demand in wireless image and video transmission requires high data rate and reliable transmission of multimedia stream over severe channel conditions. However, such services are bandwidth intensive [1], and the time-varying wireless channels are shared by a huge number of users. These problems along with the total power consumption limitations impose challenges in maintaining high quality received videos and images.

Investigations have shown that a large portion of the licensed frequency spectrum is either entirely unused or seldom being utilized [2]. Naturally, a more effective utilization of the spectrum is desirable. Cognitive radio (CR) has emerged as a powerful scheme for unlicensed devices to use licensed bands under certain conditions, and the scheme was approved by the Federal Communication Commission (FCC) in 2002 [3].

Two dynamic spectrum access mechanisms have been proposed in literature based on the coexistence with respect to the primary user (PU): i) underlay spectrum access and ii) overlay spectrum access [4]. In the underlay spectrum access, the PU and secondary user (SU) can coexist in the same spectral band, while in the overlay CR the SUs can only access the unused licensed channel resources assigned to PUs, when the PU has no data to transmit [5]. In this paper the overlay CR is chosen as the access mechanism, which requires some flexibilities in the spectrum shape of the transmitted signals. Orthogonal frequency division multiplexing (OFDM) offers this kind of flexibilities by filling the spectral gaps without interfering with PUs [6]. However, a major problem with this kind of applications is the mutual interference between PU and SU. The interference to PU cannot exceed its tolerable collision level, which is dependent on the primary service type. This has been widely studied in [7], [8]. A considerable amount of literature has been published on subcarrier and power allocation in OFDM-based CR [8], [9], [10].

The above works, however, have mostly concentrated on the maximization of the throughput of the system under resource constraints with the assumption that all information bits in the transmitted bitstream are equally important. Applying these schemes to the transmission of scalable multimedia bitstreams results in poor resource utilization [11]. Therefore, several attempts have been made to address this issue. In [12], the authors optimized video streaming to exploit more channel resources for SU by developing a flexible sensing-transmission scheme. In [7], a cross-layer quality-aware resource allocation algorithm to optimize OFDM access-based (OFDMA-based) CR network performance was proposed, by considering the imperfect channel state information between the SU and PU. It also considers quality-aware resource allocation using the H.264/AVC standard. The study in [13] proposed a bit-error-rate-driven (BER-driven) resource allocation for scalable bitstreams over OFDMA systems, which can be applied to the CR networks with some changes in the considered constraints.

As suggested in [14], removing the redundancies via data compression leads to bandwidth saving, but it increases the sensitivity of the data to transmission errors. In [15], a new hyperspectral band ordering algorithm is presented that improves the compression performance. JPEG 2000 standard has a quality progression feature, which can improve the received image quality progressively as more data from different quality layers are received [16]. In order to improve the transmission performance, the important parts of the bitstream (lower layers) should be transmitted with more protection than the less important parts. This is known as unequal error protection [13], [17], [18], [19], [20], [21], [22], [23].

In this paper, we improve the received image quality in a CR system by taking advantage of the scalable bitstream and unequal power allocation in two stages. The first stage optimizes the power allocated to the JPEG 2000 bitstream at the coding pass (CP) level to minimize the total received distortion. The second stage employs subcarrier allocation, adaptive modulation, and power adjustment to meet the interference requirements, based on channel conditions, and at the same time keeps the same throughput for the system. This strategy is expected to enhance the image quality, since important parts of the image will be transmitted more reliably.

Section snippets

System model

The overall system block diagram of the proposed scheme is shown in Fig. 1, where the first block encodes the input image into the JPEG 2000 format. In the JPEG 2000 coding process, first the raw image is partitioned into a number of rectangular non-overlapping blocks which are referred to as tiles. Then a discrete wavelet transform (DWT) is applied to each tile to decompose it into sub-bands at different resolution levels. The first decomposition level has four sub-bands, LL1,LH1,HL1, and HH1.

Unequal power allocation

Eq. (1) is a non-linear optimization problem. Since the objective function in Eq. (1) is not a convex function, Gradient-based methods such as Lagrange multipliers, are not appropriate for finding the optimum solution of this problem and they may find the local optimums instead. Therefore, to solve the UPA problem, simulated annealing (SA) method is used to locate an acceptable approximation of the global minimum in a large search space, as suggested in [27]. Due to the large number of CPs and

Channel allocation and power adjustment

The purpose of the channel allocation block in Fig. 3 is to further protect the important JPEG 2000 bits by transmitting them over sub-channels with higher quality. On the channel allocation block, sub-channels with better conditions are assigned to the important layers of JPEG 2000 image. It can be seen from Eq. (7) that by increasing hSP, the imposed interference on the PU increases. Therefore, the ratio between hSS and hSP can be an acceptable representation of the channel condition. This

Simulation results

To assess the proposed method, a grayscale image of Lenna, with the size of 512×512 and 8 bit/pixel is used for transmission and the Kakadu software is utilized as the JPEG 2000 image coder. The image is processed with one level of decomposition and is divided into 64×64 CBs and 128×128 precincts. The final image has 3 layers with the same size. For the baseline scenario, 16 sub-carriers, one PU, and one SU are considered. Table 5 shows the selected values for the model parameters based on the

Conclusion

In this work, resource allocation algorithm is applied to enhance the quality of transmitted JPEG 2000 codec images over OFDM-based CR by considering the power and interference constraints. An optimization algorithm is used to allocate unequal power to the bits of image bitstream based on their impact on the image quality. The important bits in lower layers are further protected by applying subchannel allocation on top of UPA algorithm. The developed algorithm was observed to improve the SU’s

References (29)

  • NgoD.T. et al.

    Resource allocation for OFDMA-based cognitive radio multicast networks with primary user activity consideration

    IEEE Trans. Veh. Technol.

    (2010)
  • DashtiM. et al.

    Radio resource allocation for orthogonal frequency division multiple access-based underlay cognitive radio networks utilising weighted ergodic rates

    IET Commun.

    (2012)
  • JangJiho et al.

    Transmit power adaptation for multiuser OFDM systems

    IEEE J. Sel. Areas Commun.

    (2003)
  • ZhuH.

    Radio resource allocation for OFDMA systems in high speed environments

    IEEE J. Sel. Areas Commun.

    (2012)
  • Cited by (6)

    • Multimedia communication over cognitive radio networks from QoS/QoE perspective: A comprehensive survey

      2020, Journal of Network and Computer Applications
      Citation Excerpt :

      Thus, SUs need to periodically monitor PUs activity on the desired spectrum bands, transmit their data over WSs without any harmful interference in an opportunistic manner. In overlay mode, sophisticated signal processing and coding, such as dirty paper coding (DPC), is used by CR to maintain or improve the communication of PUs while capturing extra spectrum bands for SUs' communications (Javadi et al., 2017). In this mode, SUs are required to be aware of transmitted data sequences (messages), channel gains as well as the encoding procedure of the sequences (codebooks).

    • A multiple description scalable coding scheme for adaptive video streaming over next generation cellular networks

      2019, Digital Signal Processing: A Review Journal
      Citation Excerpt :

      A general cross layer model is proposed in [8] to ensure end-to-end multimedia content delivery in realistic cognitive radio contexts, characterized by random primary users appearance and channel impairments as well. A recent work [10] implements an unequal power allocation mechanism in an IEEE 802.11a context, the coded bits with the largest contribution to the overall image quality are granted higher power levels. The optimal solution is obtained by minimizing the image distortion while properly complying with the interference constraint of the primary system.

    • Hardware implementation of digital image coding using fdct

      2018, Journal of Theoretical and Applied Information Technology
    View full text