\documentclass{beamer} %\usepackage[default]{comfortaa} \usepackage{mathtools} \usepackage{verbatim} \usepackage{geometry} %\renewcommand{\familydefault}{\sfdefault} \usefonttheme{structuresmallcapsserif} \usefonttheme[onlymath]{serif} \usetheme{Hannover} \usecolortheme{sidebartab} %\setbeamerfont*{frametitle}{size=\normalsize,series=\bfseries} \setbeamertemplate{navigation symbols}{} \setbeamertemplate{items}[rectangle] \newcommand\Fontvi{\fontsize{9}{8}\selectfont} \usepackage{tikz} \usetikzlibrary{arrows,decorations.pathmorphing,backgrounds,fit,positioning,shapes.symbols,chains} \setbeameroption{hide notes} %\setbeameroption{show notes} %\setbeameroption{show only notes} \setbeamertemplate{note page}[compress] \title[Crowncom13] {Sense in Order: Channel Selection for Sensing in Cognitive Radio Networks} \author[] {Ying Dai, Jie Wu} \institute[temple] {Department of Computer and Information Sciences, Temple University} \date[] {} \begin{document} \begin{frame} \titlepage \end{frame} \section{Introduction} \begin{frame} \frametitle{Motivation} \begin{itemize} \item Spectrum sensing is one of the key phases in Cognitive radio networks (CRNs). \vspace{0.2cm} \pause \item Before data transmission happens, each node (secondary user) needs to find one available channel. \vspace{0.2cm} \pause \item If the channel is unavailable, it needs to adjust its parameters and switch to sense another channel. \end{itemize} \note{Cognitive radio networks are a promising solution to the spectrum congestion problem. Primary users in CRNs are privileged users, for whom transmission is free from interference. Each node or secondary user, is capable of sensing the available channels and can make opportunistic use of them without causing interference to primary users. During the process, one of the most important phases is the spectrum sensing. Each time a secondary user needs to find one channel for transmission, it will pick one channel for sensing. If the channel is unavailable, it needs to adjust its parameters and switch to sense another channel. } \end{frame} \begin{frame} \frametitle{Motivation} An example: \pause \begin{center} \includegraphics[scale=0.80]{1.eps} \end{center} \pause \emph{\textbf{Q}}: How to increase the efficiency for spectrum sensing? \note{For example, in this figure, there is a pair of primary users, TX and RX. The secondary user $u$ is in the interference range TX, who is using the channel $m_1$ to send data to RX. There is a total of three channels ($m_1$, $m_2$, $m_3$) in the network. If $u$ needs to use one channel, it will pick one channel from the three channels. Because there are no differences among the three channels from $u$'s point of view, it is possible that $u$ will pick $m_1$ for sensing. Then, after $u$ finds out that $m_1$ is unavailable based on the sensing results, it must switch to another channel to sense again. Here comes the question: How to increase the efficiency for spectrum sensing? } \end{frame} \begin{frame} \frametitle{Motivation} \begin{itemize} \item Before spectrum sensing, choose the channel that is more likely to be available for sensing. \vspace{0.2cm} \pause \item This is practical with the help of nodes nearby. \pause {\small{~~~~$\circ$ For example, in previous figure, node $u$ is likely to know which channels are more likely to be available by overhearing some information provided by $v$ and $w$.}} \end{itemize} \note{Here is one way to increase the efficiency for spectrum sensing. Before spectrum sensing, instead of treating all channels the same and randomly picking one channel to sense, each node selects the channels that are more possible to be available to sense. This is practical if we can use the help from nodes nearby. Consider the previous figure, for example, node $u$ is possible to know which channels are more possible to be available by overhearing some information provided by $v$ and $w$. } \end{frame} \section{System Model} \subsection{overview} \begin{frame} \frametitle{Overview} \begin{itemize} \item How to choose a channel for sensing for each node at the beginning: \pause {\small{~~~~$\circ$ ``Pre-phase'' of spectrum sensing: it happens before the spectrum sensing}} \pause \item We propose a sense-in-order (SIO) model for the pre-phase problem: \pause {\small{~~~~$\circ$ The order is determined before the spectrum sensing, and is maintained as a list by each node.}} \pause \item Each looks up the list and selects a channel for sensing. \pause {\small{~~~~$\circ$ Each node knows the order to sense, which results in a reduction of switches among channels during spectrum sensing.}} \end{itemize} \note{ We consider how to choose a channel for sensing for each node at the beginning, so that the probability of switching to sense another channel is reduced. We call this ``Pre-phase'' of spectrum sensing, which happens before the spectrum sensing. We propose a sense-in-order (SIO) model, which provides each node with an order for spectrum sensing. The order is determined before the spectrum sensing, and is maintained as a list by each node. When a node needs to find a channel for data transmission, it can look up the list and select a channel that has a higher probability of being available for sensing. Each node knows the order to sense, which results in a reduction of switches among channels during spectrum sensing. } \end{frame} \subsection{Problem Formulation} \begin{frame} \frametitle{Problem Formulation} \begin{itemize} \item A channel is sensed as available if and only if it is neither occupied by primary users nor secondary users. \pause \vspace{0.2cm} \item We define the cost $C_v$ of each node $v$ during the spectrum sensing as the number of switches among channels that are needed until an available one is found. \pause \vspace{0.2cm} \item \emph{\textbf{Objective: }} Provide an order of channels for sensing so that the cost during the spectrum sensing phase is minimized: $\text{Min } \sum_{v\in N}C_v$. \end{itemize} \note{In our model, a channel is sensed as available if and only if it is neither occupied by primary users nor secondary users. When the channel is sensed as available, the secondary users can access the channel and use that for transmission. Otherwise, secondary users need to switch to another channel for sensing, which costs more energy and delay. We assume that the delay and energy cost of each switch from one channel to another channel for sensing is a constant. Therefore, we define the cost $C_v$ of each node $v$ ($v \in N$) during the spectrum sensing phase is calculated as the number of switches among channels that are needed until an available one is found. Our objective is to provide an order of channels for sensing so that the cost during the spectrum sensing phase is minimized.} \end{frame} \subsection{State Transition Diagram} \begin{frame} \frametitle{Sense-in-order Model} \begin{itemize} \item Each node senses the channel when it needs a channel for transmission, and broadcasts the sensing results through common control channel. \vspace{0.2cm} \pause \item If the node finds an available channel, it will access that channel. \vspace{0.2cm} \pause \item The node will also broadcast when it accesses and when it quits that channel. \end{itemize} \note{We assume that there is a common control channel for exchanging control information. To simplify our mode, we assume that there is no loss of the signal transmission over the CCC. Each node broadcasts its sensing results through the CCC. And each node does not sense the channel until it needs a channel for transmission. Thus, if the node finds an available channel, it will access that channel. It will also broadcast when it accesses and when it quits that channel.} \end{frame} \begin{frame} \frametitle{Sense-in-order Model} The broadcast information can be implemented using the following three signals: \pause \vspace{0.2cm} \begin{itemize} \item $PO_m$: channel $m$ is occupied by primary users; \vspace{0.2cm} \pause \item $SO_m$: channel $m$ is free from primary users, but is occupied by the secondary user who sent this signal; \vspace{0.2cm} \pause \item $SF_m$: Secondary user finishes transmission and quit from channel $m$. \end{itemize} \note{Since there are several types of information need to be broadcast, we use different signals to implement this scheme. There are three types of signals: $PO_m$: channel $m$ is occupied by primary users; $SO_m$: channel $m$ is free from primary users, but is occupied by the secondary user who sent this signal; $SF_m$: Secondary user finishes transmission and quit from channel $m$. If channel $m$ is occupied by secondary users, node $u$ will avoid sensing that channel since $SO_m$ was received previously by $u$ until the $SF_m$ is received.} \end{frame} \begin{frame} \frametitle{Sense-in-order Model} \begin{itemize} \item Based on the received signals, a node $v$ is able to identify four different states, $S = \{S_i, 1 \le i \le 4\}$, for a channel $m$. \vspace{0.2cm} \pause \item We use $$ to indicate that channel $m$ is in state $S_i$: \vspace{0.2cm} \pause {\small{~~~~$\circ$ $$: $m$ is occupied by primary users;}} {\small{~~~~$\circ$ $$: $m$ is not occupied by primary users, but is occupied by the secondary user;}} {\small{~~~~$\circ$ $$: the secondary user previously using $m$ has finished transmission and quit from $m$;}} {\small{~~~~$\circ$ $$: no signal is received about $m$.}} \end{itemize} \note{Read this slide.} \end{frame} \begin{frame} \frametitle{Sense-in-order Model} \begin{itemize} \item The four states are maintained on each node itself. \pause \item For $$, node $v$ is not sure about whether the primary users have finished transmission on $m$ if no other sensing results are received from other nodes. \vspace{0.2cm} \pause \item For $$, node $v$ should avoid sensing $m$ until $v$ receives the signal $SF_m$. \vspace{0.2cm} \pause \item For $$, node $v$ should assign higher probabilities for selecting $m$ to sense. \vspace{0.2cm} \pause \item For $$, $v$ is not sure about the availability of $m$ either. \end{itemize} \note{Here are some explanations of the four states. Read the slide.} \end{frame} \begin{frame} \frametitle{Sense-in-order Model} \begin{itemize} \item Each node changes among the four states based on the signal it receives. \pause \begin{center} \includegraphics[scale=0.80]{4.eps} \end{center} \end{itemize} \note{ The initial state of each channel is $S_4$. Each state is updated based on the most recent signal in the time domain. Since primary users have higher privileges on each channel, $PO_m$ could be received no matter what the previous state is. State $S_3$ can only be reached from $S_2$, since we assume that there is no packet loss. Each node will mark the channel as $S_2$ after receiving $SO_m$, and will not update the state as $S_3$ or $S_4$ before $SF_m$ is received. Thus, there is no valid time expiration issue for state $S_2$.} \end{frame} \subsection{Channel Selection Algorithm} \begin{frame} \frametitle{Sense-in-order Model} \begin{itemize} \item How does each node define preferences on different channels: \vspace{0.2cm} \pause \item Each node divides the whole channel set into four (at most) different subsets, based on the state of each channel. \pause {\small{~~~~$\circ$ For node $v$, the whole channel set $M$ is divided into four subsets $M_v(S_i)$, $1 \le i \le 4$.}} {\small{~~~~$\circ$ If channel $m \in M_v(S_i)$, channel $m$ is identified as state $S_i$ by node $v$. }} \pause \vspace{0.2cm} \item The probability of each channel to be chosen for sensing is: {\tiny{ $$p_v^m = \begin{cases} \frac{t_m}{\sum_{m_0 \in M_v(S_1)} t_{m_0}} \times P_v(S_1) & m \in M_v(S_1)\\ \\ 0 & m \in M_v(S_2)\\ \frac{T - t_m}{\sum_{m_0 \in M_v(S_3)} (T - t_{m_0})} \times P_v(S_3) & m \in M_v(S_3)\\ \\ \frac{P_v(S_4)}{|M_v(S_4)|} & m \in M_v(S_4) \end{cases}.$$}} \end{itemize} \note{Since each node stores the state for every channel, it needs to define preferences on different channels when it needs to select one channel for sensing. Each node divides the whole channel set into four (at most) different subsets, based on the state of each channel. For node $v$, the whole channel set $M$ is divided into four subsets $M_v(S_i)$, $1 \le i \le 4$. If channel $m \in M_v(S_i)$, channel $m$ is identified as state $S_i$ by node $v$. We first determine the probability of each of the four sets of channels to be chosen. Then we decide the probability of a single channel to be chosen. The probability is as shown here. The details are explained in our paper.} \end{frame} \begin{frame} \frametitle{Sense-in-order Model} The overall structure of our algorithm for a node $v$ is: \begin{itemize} \item $v$ updates the state of each channel based on the received signal; \pause \vspace{0.2cm} \item When $v$ needs to transmit data, it calculates the probability of each channel to be chosen and selects one channel to sense until it finds an available one; \pause \vspace{0.2cm} \item $v$ shares its sensing results with others and sends out the corresponding signal when it accesses and quits that channel. \end{itemize} \note{read this slide} \end{frame} \section{Simulation} \begin{frame} \frametitle{Simulation Results} We evaluate our algorithm performance by varying different parameters, including both network parameters and algorithm parameters. \begin{center} \includegraphics[scale=0.60]{6.png} \end{center} \note{We evaluate our algorithm performance by varying different parameters, including both network parameters and algorithm parameters.} \end{frame} \section{Conclusion} \begin{frame} \frametitle{Conclusion} \begin{itemize} \item We consider the pre-phase of spectrum sensing, which focus on how to choose a channel for sensing for each node in cognitive radio networks (CRNs). \vspace{0.2cm} \item We propose an SIO model, which constructs a state transition diagram and a corresponding algorithm for each node to calculate the probability of each channel being chosen for sensing. \vspace{0.2cm} \item Extensive simulation results testify the efficiency of our model. \end{itemize} \end{frame} \begin{frame} \begin{center} \LARGE \emph{Thank you!} \end{center} \end{frame} \end{document}