Intelligent Security Management and Control in the IoT. Mohamed-Aymen Chalouf

Чтение книги онлайн.

СКАЧАТЬ Impact on performance of the number of IoT devices simultaneously attempting access

      The idea introduced in this chapter is quite simple, because it involves calculating a blocking factor. Nevertheless, a good implementation would require a good knowledge of the number of terminals ready to attempt access, so as to deduce from this the probability of the optimal blocking. This information is unfortunately not available in the network.

      To solve this problem, two significant challenges should be taken into account: (1) designing an access control strategy for dynamic generation of the blocking factor and (2) estimating the number of devices simultaneously attempting to access the network.

      In this chapter, we tackle these questions using an estimator that was suggested in earlier research (Bouzouita et al. 2019). Since this estimation is very noisy, we exploit the potential of the most advanced reinforcement learning techniques, to take account of this complex reality (the state of the network is not observable) and deduce a sub-optimal control strategy. More especially in this chapter, we use the deep reinforcement learning algorithm Twin Delayed Deep Deterministic policy gradient algorithm (TD3) (Fujimoto et al. 2018) to produce the optimal blocking factor from these past estimations (Hadjadj-Aoul and Ait-Chellouche 2020).

The remainder of this chapter is organized as follows. Section 2.2 briefly presents the NB-IoT standard considered in our proposal. Section 2.3 gives an overview of the main congestion control techniques in IoT networks, and more especially cellular IoT networks. Section 2.4 describes the model for IoT terminals to access the network. Section 2.5 describes the suggested control solution, based on the TD3 algorithm, adapted to solve the problem of calculating the blocking factor. Section 2.6 describes the environment for simulating the suggested approach and shows its effectiveness compared to the existing approach. Finally, the chapter ends with a summary recapitulating the main advantages and achievements of the system suggested in section 2.7.

2.2. Fundamentals of the NB-IoT standard

      NB-IoT is a cellular access technology of the Low Power Wide Area Network (LPWAN) type, specified in the document “Rel-13” by the 3GPP (3GPP 2016). Improvements to it were then provided in the documents “Rel-14”, “Rel-15” and “Rel-16” (Mwakwata et al. 2019). Several key players in the industry, such as Ericsson, Nokia, Intel or Huawei, have shown a keen interest in this standard and have broadly participated in its standardization (Bicheno 2015).

      NB-IoT technology is also presented as a promising candidate for covering one of the three main pillars of 5G, that is the massive Machine Type Communication (mMTC) (3GPP 2018a; Narayanan et al. 2018). In fact, the ITU has defined different requirements in terms of the density of the IoT terminals, battery life, coverage, price and mechanisms and features supported, to which the suggested technologies must respond to implement mMTCs (Kafle et al. 2016). In this direction, the 3GPP has suggested several mechanisms and NB-IoT features to meet these demands through different releases (Mwakwata et al. 2019). The options for integrating the NB-IoT with the core 5G network and its coexistence with the other services suggested in the framework of this standard are also studied by the 3GPP (3GPP 2019a). A better use of physical resources via the virtualization of network functions (NFV) or the SDN, as discussed in Migabo et al. (2020), can also facilitate this integration.

      2.2.1. Deployment and instances of use

Many operators worldwide chose the NB-IoT standard from when it was standardized in 2016. Indeed, its properties, inherent to cellular technologies such as flexibility, adaptability, “over-the-air” service updates and cost containment, have of the three main pillars of 5G, that is the massive Machine Type Communication (mMTC) (3GPP 2018a; Narayanan et al. 2018). In fact, accelerated adoption of this standard. According to a study by GSMA Intelligence, the total number of cellular IoT connections should reach 3.5 billion by 2025 (GSMA 2020). According to the same study, 94 commercial NB-IoT networks were listed in 2020. These latter are mainly located in China, where the standard is promised by the manufacturer Huawei. In Europe, Vodafone is the main operator promoting this standard in France, the operator SFR has integrated NB-IoT to its provision since last year.

      This adoption of NB-IoT, by operators on the one hand and by manufacturers on the other, has occasioned many applications (Huawei 2017; Ray 2017), particularly in the Industry 4.0 domains, smart measurements, connected cities, monitoring and surveillance, smart agriculture and connected farms, or health.

      2.2.2. Transmission principles

      As its name indicates, NB-IoT is based on signal transmission over a narrow band. In fact, its spectral occupation is only 180 kHz, which is equivalent to an LTE Physical Resource Block (PRB). Because of this, it results in very low flow rates (at most 250 kbit/s), but these are sufficient for the applications for which this protocol is designed.

      NB-IoT technology relies on LTE from which it inherits numerous features and mechanisms, especially at the level of physical and MAC layers. It also re-uses the same numerologies, channel coding, interleaving, etc. This has made it possible to reduce, on the one hand, the time needed to specify this standard and, on the other hand, the costs of developing the NB-IoT devices. However, since the bandwidth used is very narrow, modifications were needed to permit the primary objectives of this standard, that is, the massive use of long-range connections at reduced complexity and cost (Flore 2016).

      2.2.2.1. Modes of deployment

      As illustrated in Figure 2.2, NB-IoT can be deployed in three different modes: (1) in-band, within the LTE frequency band by substituting it for a PRB; (2) guard-band, by making use of unused spectrum resources, outside the traditional LTE frequency band; or finally (3) standalone, on an independent carrier. In this latter case, GSM frequencies are most often marked for deployment (Wang et al. 2017).

      Figure 2.2. NB-IoT deployment modes

      The three deployment modes cited above are transparent for non NB-IoT terminals. These consider the neighboring NB-IoT communications as noise.

       2.2.2.2. Physical layer

      On the downlink, transmissions are based on OFDM. Each OFDM symbol occupies 12 sub-carriers each of 15 kHz, thus occupying the equivalent of an LTE PRB (180 kHz). Also, in a similar way to LTE, each frame is composed of 10 sub-frames, each of two slots. The durations of the frame, the sub-frame and the slot are 10 ms, 2 ms and 0.5 ms, respectively. Each terminal operates on a Resource Block (RB) formed of seven consecutive OFDM symbols over the whole of the frequency domain. Concerning modulation, only Quadrature Phase-Shift Keying (QPSK) is supported.

      On the uplink, transmissions are based on SC-FDMA (Single Carrier-Frequency Division Multiple Access). Two configuration modes are possible: single tone or multi-tone. When single tone is used, the two numerologies 15 kHz and 3.75 kHz are possible. The RU radio resource unit allocated to the terminals corresponds СКАЧАТЬ